Although stressful life events have been found to be important markers of depression onsets in psychiatric settings, stressful events have rarely been examined in primary care settings. The overriding objective of the proposed research is to examine the impact of recent stressful events on physician detection, treatment, and outcome of depression in primary care settings. It is proposed that patients and physicians fail to acknowledge depression-related symptoms because depression following stressful life events is construed as a "normal" reaction to adverse circumstances. Stressful life events may thus constitute an important factor in accounting for low rates of patient disclosure of symptoms, as well as low rates of physician diagnosis and treatment of depression in primary care. However, recent outcome studies raise the possibility that patients and physicians may be correct in assuming that depression following a stressful event does not invariably require treatment. That is undetected depressed patients exhibit symptomatic improvement after a few months without treatment. In short, stressful life events may produce "distress reactions" that meet criteria for depression but remit without treatment, in a manner consistent with patients' and physicians' expectations. The proposed study will use semi-structured diagnostic interviews to assess depression in patients screened in the waiting rooms of their primary care physicians. A checklist screening measure and in-depth, contextual interviews will be used to provide valid assessments of the stressfulness of life events. Physician detection of depression will be assessed at the time of the screening encounter and at follow-up, which will occur after 4 months and involve the assessment of mood-related symptoms and major depression in patient groups who both report and fail to report stressful life events. It is predicted that in the context of a recent stressful life event, depressed primary care patients will not mention their mood to their physicians, who in turn will not diagnose or treat depression. However, at follow-up, these patients are expected to show greater improvements in depressive symptoms than patients who had not experienced a stressful event prior to the onset of their depression. Thus, this design will begin to specify the characteristics of depressions that require detection and treatment in primary care settings versus the characteristics of depressions that more typically improve spontaneously. Finally, the findings generated from this research program may well serve to pinpoint detection and treatment efforts and replace current blanket and unfocused efforts to detect and treat depression in primary care. Data from this study are expected to be published and will provide pilot data for future RO1 grant submissions as well as demonstrate the logistical capabilities and community partnerships necessary for studying depression in primary care settings.